We already know that it is very difficult to obtain an "isolated field" for direct bonding during the surgical exposure of unerupted teeth. The aim of this in-vitro study is to simulate the clinical situation of forced eruption and to evaluate the tensile strengths of preligatured button with several types of contamination which can happen during the surgical exposure of unerupted teeth. Four orthodontic direct bonding systems were used. ($Ortho-One^{TM}$, $Rely-a-Bond^{(R)}$, $Ortho-Two^{TM}$, Phase $II^{(R)}$) Each material was divided into four groups(n=20) : Group 1. (Control, no contamination), Group 2. (Rinse etching agent with saline instead of water), Group 3. (Blood contamination of etched surface for 30 seconds), Group 4. (Blood contamination of primed surface for 30 seconds) 320 bovine anterior permanent teeth were divided into the above mentioned 16 groups. Enamel surface was flattened and ground under water coolant. Pre-ligatured buttons were prepared to the same form. (Cut 0.25 ligature wire 10 cm in length. Twist the ligature wire 30 times clockwise. Mark the wire 15mm and 35mm points from button. Make a loop sticking two points together and twist the loop 6 times counterclockwise.) The bonded specimens were stored at $37^{\circ}C$ saline solution for 3 days. Then the tensile strength of each sample was measured with Instron universal testing machine, crosshead speed of 0.5mm/min. The following results were obtained: 1. As compared to control groups (Group 1) of each material, Rely-a-Bond had a significantly lower mean tensile strengths than other material. (p<0.01) 2. In Group 2. of Ortho-One and Rely-a-Bond, the mean tensile strengths decreased about 7.7% and 11.1%, respectively with statistical significances. (p<0.05) 3. In Group 2. of Ortho-Two and Phase II, the mean tensile strengths did not decrease. 4. In Group 3. of Ortho-One, Rely-a-Bond, Ortho-Two, and Phase II, the mean tensile strengths decreased about 60.8%, 56.1%, 60.2%, and 46.0%, respectively with statistical significances. (p<0.01) 5. In Group 4. of Ortho-One and Rely-a-Bond, the mean tensile strengths did not decrease. 6. In Group 4. of Ortho-Two and Phase II, the mean tensile strengths were decreased about 20.95% and 22.28%, respectively with statistical significances. (p<0.01) There were formations of a hump shaped mass from bonding resin under blood contamination which disturbed direct bonding procedure. According to Reynolds, the proper bond strength for clinical manipulation should be at least 45N or about 4.5Kg.F. According to these results, it can be concluded that Ortho-One could be used during surgical exposure of unerupted teeth. In any case, blood contamination of the etched surface should be avoided, but the blood contamination of primed surface of Ortho-One may not decrease bond strength. Just 'blowing-out' is enough to remove blood from primed surface of Ortho-One. You can verify the clean surface of the primer of Ortho-One after blowing out the blood contamination.